Molecular Sets (MOSES): A Benchmarking Platform for Molecular Generation Models

Abstract

Generative models are becoming a tool of choice for exploring the molecular space. These models learn on a large training dataset and produce novel molecular structures with similar properties. Generated structures can be utilized for virtual screening or training semi-supervized predictive models in the downstream tasks. While there are plenty of generative models, it is unclear how to compare and rank them. In this work, we introduce a benchmarking platform called Molecular Sets (MOSES) to standardize training and comparison of molecular generative models. MOSES provides training and testing datasets, and a set of metrics to evaluate the quality and diversity of generated structures. We have implemented and compared several molecular generation models and suggest to use our results as reference points for further advancements in generative chemistry research. The platform and source code are available at https://github.com/molecularsets/moses.

Artificial intelligence in early drug discovery enabling precision medicine

“…Despite the increase of new AI-based small molecule design approaches, a common set of validation criteria has not yet emerged. Two recent benchmarking platforms addressed these issues [207,208]. Polykovskiy et al [208] evaluated generative models based on their capability to produce structures similar to the training set, while Brown et al [207] probed the models' ability to replicate the physicochemical property distributions of a reference set, and to generate novel molecules that optimize single or multiple criteria jointly.…”

Schistosomiasis Drug Discovery in the Era of Automation and Artificial Intelligence

“…Despite the innovation of generative models, the novelty and accessibility of generated molecules must be evaluated (372)(373)(374). Gao and Coley (375) observed that generative models can produce infeasible molecules even with good performance in benchmarks.…”

Paradigm Shift: The Promise of Deep Learning in Molecular Systems Engineering and Design

“…Given the growing interest in generative modeling from the deep learning community, applications of more complex and hybrid approaches have also been proposed (Griffiths and Hernandez-Lobato, 2020;Maziarka et al, 2020). Two benchmarking platforms have been developed for evaluating distinct elements of the representations learned by generative models, including validity, novelty, diversity, and uniqueness (Brown et al, 2019;Polykovskiy et al, 2020). The capability to learn a domain-invariant molecular representation on different scales of complexity to generative valid and novel molecules remains a key limitation.…”